The other day I was working with TurnItIn’s Originality Report feature because my professor wanted to see that I was using it to examine my drafts. I don’t mind using tools to help me write; I tend to go from hyper-technological to just-a-terminal and everything in between depending on how I am feeling at the moment. My university provides a “draft” bucket I can use for each of my written assignments, designed to give me an Originality Report on my draft so that I can ensure my final paper that I turn in to the assignment bucket is around or less than about 10%.
After playing around with the tool for a while, I noticed that some of my draft sections were coming back as mostly unoriginal. I thought this was odd, seeing as how I rarely if ever use direct quotes from anything. Upon further investigation I realized that the report was showing me pull quotes from articles that I have published around the internet. This got me thinking.
How is TurnItIn knowing whether a report is original or not? Surely there must exist some sort of database of papers that it draws from, probably a huge repository of all the papers that have ever been pushed to the service. Then there’s the internet, of course, and I’m sure they do relevance searches with all the popular web caching services. But this leaves out a huge chunk of academic text, don’t you think?
Oh, which chunk? Microfiche. Yes, that’s right. The hundreds of thousands of hundreds of thousands (yes, that was intentional) of academic work that only exists in digital form because some poor volunteer students scanned them into a digital repository. Interestingly enough, these works exist as image snapshots – not as digital text – and unless TurnItIn was going to OCR-scan pterabytes of data for each Originality Report I think they’ll probably just skip over the microfiche archives.
Thinking that I couldn’t possibly have discovered a gold mine in breaking TurnItIn, I decided to verify that my suspicions were ill-founded. For the sake of science, I copied three full pages of someone’s dissertation from microfiche and then ran it through TurnItIn’s originality report. Can you guess what happened? Almost 100% original.
Now, I know what you’re thinking. “Who in their right mind is going to copy some microfiche text that is over forty years out of date?” A cheater, that’s who. And since TurnItIn and similar serves market themselves as the end-all be-all for academic plagiarism detection, this gigantic loophole presents a giant monkey wrench in the whole system.
I’ll admit that I have on occasion put too much faith into these originality reports when grading student papers. I’m sure I’m not the only one who has done it, too. What I’ve discovered here is that the tools that we thought were going to help make our lives easier still have the same workarounds that crappy students will find out about eventually.
Perhaps that’s the solace we can take in this, though. Crappy students are going to do crappy things, like copy microfiche papers in order to make it past digital plagiarism detection devices. In the end they’ll probably pass their classes – hell, they’ll probably get a degree, too – but just like everyone else complaining that they can’t find a job after they got their degree, they’ll quickly realize that, just like our greatest strength, their work ethic is going to be their greatest downfall.